Fault Diagnosis of Planetary Gearbox using ICEEMDAN and SVM
-
摘要: 针对行星齿轮箱复合故障准确分类问题,应用了改进自适应噪声完备集合经验模态分解(ICEEMDAN)和支持向量机(SVM)相结合的故障诊断方法。首先,将行星齿轮箱的不同故障信号分别进行ICEEMDAN分解,得到各阶内禀模态函数(IMF); 其次,利用各阶IMF分量与原信号的相关性大小,剔除虚假的IMF分量; 最后,以优选IMF分量的多尺度模糊熵均值作为特征向量,输入到多分类SVM中进行故障分类,分类准确率高达100%,实验结果证明了该方法的可行性。
-
关键词:
- 改进自适应噪声完备集合经验模态分解 /
- 频域互相关 /
- 多尺度模糊熵 /
- 支持向量机 /
- 行星齿轮箱故障
Abstract: Aiming at the problem of accurate classification of compound faults of planetary gearboxes, a fault diagnosis method combining improved adaptive noise complete set empirical mode decomposition (ICEEMDAN) and support vector machine (SVM) is proposed in this study. First, the different fault signals of the planetary gearbox are decomposed by ICEEMDAN to obtain the intrinsic mode function (IMF) of each order. Second, the correlation between the IMF component of each order and the original signal is used to remove the false IMF component. Finally, the multi-scale fuzzy entropy average value of the preferred IMF component is used as the feature vector and input into the multi-class SVM to accomplish fault classification. The classification accuracy is as high as 100%. The experimental results prove the feasibility of this method. -
表 1 行星齿轮箱4种工况的部分特征向量
工况 IMF1 IMF2 IMF3 IMF4 IMF5 IMF7 1 0.169
0.168
0.169
0.160
0.1650.233
0.235
0.224
0.232
0.2250.356
0.288
0.373
0.311
0.3540.566
0.575
0.550
0.571
0.5270.421
0.447
0.416
0.453
0.4630.362
0.374
0.343
0.354
0.3662 0.094
0.091
0.085
0.096
0.0810.143
0.145
0.140
0.148
0.1450.343
0.310
0.333
0.287
0.3090.426
0.421
0.386
0.447
0.4010.453
0.542
0.554
0.510
0.4770.374
0.350
0.379
0.306
0.3573 0.056
0.054
0.059
0.056
0.0590.139
0.125
0.132
0.129
0.1210.348
0.287
0.311
0.312
0.3180.469
0.474
0.469
0.479
0.4560.462
0.453
0.554
0.469
0.5210.345
0.371
0.352
0.382
0.3924 0.048
0.050
0.050
0.055
0.0460.163
0.151
0.146
0.157
0.1560.288
0.282
0.280
0.304
0.3040.468
0.434
0.432
0.441
0.4080.401
0.362
0.404
0.320
0.4130.356
0.290
0.370
0.384
0.304表 2 3种方法分类对比
特征提取方法 分类准确率/% 每种工况用时/s EMD
CEEMDAN
ICEEMDAN90
95
100154.465
726.077
554.869 -
[1] 饶振刚. 行星齿轮传动设计[M]. 北京: 化学工业出版社, 2007RAO Z G. Planetary gear transmission design[M]. Beijing: Chemical Industry Press, 2007 (in Chinese) [2] 丁康, 李巍华, 朱小勇. 齿轮及齿轮箱故障诊断实用技术[M]. 北京: 机械工业出版社, 2005: 1-4DING K, LI W H, ZHU X Y. Practical technology for fault diagnosis of gears and gearboxes[M]. Beijing: Mechanical Industry Press, 2005: 1-4 (in Chinese) [3] 雷亚国, 何正嘉, 林京, 等. 行星齿轮箱故障诊断技术的研究进展[J]. 机械工程学报, 2011, 47(19): 59-67 https://www.cnki.com.cn/Article/CJFDTOTAL-JXXB201119010.htmLEI Y G, HE Z J, LIN J, et al. Research advances of fault diagnosis technique for planetary gearboxes[J]. Journal of Mechanical Engineering, 2011, 47(19): 59-67 (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-JXXB201119010.htm [4] 熊国良, 甄灿壮, 张龙, 等. 基于CEEMDAN多尺度排列熵的轴承故障智能识别Fisher-GG聚类方法[J]. 噪声与振动控制, 2020, 40(6): 1-7+28 https://www.cnki.com.cn/Article/CJFDTOTAL-ZSZK202006002.htmXIONG G L, ZHEN C Z, ZHANG L, et al. Intelligent fault recognition of rolling bearings using fisher-GG clustering and CEEMDAN-based multi-scale permutation entropy[J]. Noise and Vibration Control, 2020, 40(6): 1-7+28 (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-ZSZK202006002.htm [5] TORRES M E, COLOMINAS M A, SCHLOTTHAUER G, et al. A complete ensemble empirical mode decomposition with adaptive noise[C]//2011 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). Prague, Czech Republic: IEEE, 2011: 4144-4147 [6] COLOMINAS M A, SCHLOTTHAUER G, TORRES M E. Improved complete ensemble EMD: A suitable tool for biomedical signal processing[J]. Biomedical Signal Processing and Control, 2014, 14: 19-29 doi: 10.1016/j.bspc.2014.06.009 [7] 裴振伟, 朱平. 基于ICEEMDAN-MLP的肺音信号识别研究[J]. 电子设计工程, 2021, 29(1): 96-100 https://www.cnki.com.cn/Article/CJFDTOTAL-GWDZ202101020.htmPEI Z W, ZHU P. Research on lung sound signal recognition based on ICEEMDAN-MLP[J]. Electronic Design Engineering, 2021, 29(1): 96-100 (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-GWDZ202101020.htm [8] 管一臣, 童攀, 冯志鹏. 基于ICEEMDAN方法和频率解调的行星齿轮箱故障电流信号特征分析[J]. 振动与冲击, 2019, 38(24): 41-47 https://www.cnki.com.cn/Article/CJFDTOTAL-ZDCJ201924006.htmGUAN Y C, TONG P, FENG Z P. Planetary gearbox fault diagnosis via current signature analysis based on ICEEMDAN and frequency demodulation[J]. Journal of Vibration and Shock, 2019, 38(24): 41-47 (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-ZDCJ201924006.htm [9] 姚瑞琦. 基于信息熵的直升机自动倾斜器滚动轴承故障诊断方法研究[D]. 南昌: 南昌航空大学, 2017YAO R Q. Research on fault diagnosis method of helicopter swashplate rolling bearing based on information entropy[D]. Nanchang: Nanchang Hangkong University, 2017 (in Chinese) [10] ZHANG J X, GUO Y H, SHEN Y L, et al. Improved CEEMDAN-wavelet transform de-noising method and its application in well logging noise reduction[J]. Journal of Geophysics and Engineering, 2018, 15(3): 775-787 doi: 10.1088/1742-2140/aaa076 [11] 张梅军, 唐建, 何晓晖. EEMD方法及其在机械故障诊断中的应用[M]. 北京: 国防工业出版社, 2015: 76-84ZHANG M J, TANG J, HE X H. EEMD method and its application in mechanical fault diagnosis[M]. Beijing: National Defense Industry Press, 2015: 76-84 (in Chinese) [12] 郑近德, 陈敏均, 程军圣, 等. 多尺度模糊熵及其在滚动轴承故障诊断中的应用[J]. 振动工程学报, 2014, 27(1): 145-151 https://www.cnki.com.cn/Article/CJFDTOTAL-ZDGC201401020.htmZHENG J D, CHEN M J, CHENG J S, et al. Multiscale fuzzy entropy and its application in rolling bearing fault diagnosis[J]. Journal of Vibration Engineering, 2014, 27(1): 145-151 (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-ZDGC201401020.htm [13] 柴兴亮, 刘薇娜. 基于改进CEEMDAN和多尺度模糊熵的气阀故障诊断[J]. 组合机床与自动化加工技术, 2020(10): 140-143+147 https://www.cnki.com.cn/Article/CJFDTOTAL-ZHJC202010033.htmCHAI X L, LIU W N. Gas valve fault diagnosis based on improved CEEMDAN and Multi-scale fuzzy entropy[J]. Modular Machine Tool & Automatic Manufacturing Technique, 2020(10): 140-143+147 (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-ZHJC202010033.htm [14] 周建民, 王发令, 张臣臣, 等. 基于特征优选和GA-SVM的滚动轴承智能评估方法[J]. 振动与冲击, 2021, 40(4): 227-234 https://www.cnki.com.cn/Article/CJFDTOTAL-ZDCJ202104031.htmZHOU J M, WANG F L, ZHANG C C, et al. An intelligent method for rolling bearing evaluation using feature optimization and GA-SVM[J]. Journal of Vibration and Shock, 2021, 40(4): 227-234 (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-ZDCJ202104031.htm [15] 潘峰, 唐东林, 陈印, 等. 管道腐蚀缺陷超声信号的PSO-SVM模式识别研究[J]. 机械科学与技术, 2020, 39(5): 751-757 doi: 10.13433/j.cnki.1003-8728.20190197PAN F, TANG D L, CHEN Y, et al. Ultrasonic signal pattern Recognition of pipeline corrosion defects with PSO-SVM[J]. Mechanical Science and Technology for Aerospace Engineering, 2020, 39(5): 751-757 (in Chinese) doi: 10.13433/j.cnki.1003-8728.20190197 -